In this project, I would like to define the distiction between what defines “Happiness” among different genders and marital status. This would be a crutial database which could be used as a cornerstone of development of any dating application such as Tinder and eharmony.
WordCloud will return the n number of words that was most frequently used in HappyDB. By observing these analysis first, we will be able to grasp what makes the overall people happy before we delve into different groups of people such as gender and marital status.
| 5 words | 10 words |
| 15 words | 20 words |
| 25 words | 30 words |
| 40 words | 50 words |
As we increased the number of words,
We can see there are various of values/relationships that impact the Happiness of people.
Let’s sort these values into smaller subgroups for our better analysis.
This section I used topic modeling to factor out 5 most significant topics among all the moments that make people happy.
The first topic, we can infer that these are about exercise, being motivated and early.
******
The second topic, we can infer that these are about work and job.
******
The third topic, we can infer that these are about friendship and relationship.
The fourth topic, we can infer that these are about love and family.
******
The fifth topic, we can infer that these are about entertainment.
Summary: the 5 major topics that people feel happy are being motivated, security in their jobs, friendship & relationship, family & love, and entertainment.
We are going to divide into two subtopics.
1. Difference between genders
2. Difference between marital status
We are going to analyze by using proportion difference of two groups.
Let’s look at the Comparison of Proportion graph of male and female.
Obvious result came out.
Males feel happy with their wife, girlfriend.
Females feel happy with their husband and boyfriend.
One interesting inference we can do it about how males prefer investing and gambling as the words “bitcoin” and “bet” were more towards male in the graph. Another interseting inference (or obvious) is that female feel more happy with “makeup”. There are more analysis we can do with this Comparison of Proportions graph; however, I will just list some interesting words that are more towards a certain gender.
Male: auto, video, cigarrete, alcohol, dj, and etc.
Female: closet, nap, laugh, aboard, aquarium
Summary: By analyzing these data this way, one gender would be able to find better ways to understand the opposite gender. (pun intended)
Let’s look at the Comparison of Proportion graph of marital status
Obvious result came out.
Single feel happy with their boyfriend and girlfriend
Married feel happy with their husband and wife.
One interesting inference we can do it about how married people are happy with their kids – as children, son, and daughter came out to be more towards married people. Another interseting inference (or obvious) is that single feel more happy with “text”. By which describes the situation where single people text each other a lot in order to feel happy. On the other hand, married people are happy with their anniversary since married people are more tends to keep track of their anniversary. There are more analysis we can do with this Comparison of Proportions graph; however, I will just list some interesting words that are more towards a certain gender.
Single: fiance, airbnb, alcohol, midterm
Married: daycare, car, family
Summary: By analyzing these data this way, one can analyze the gap between how single and married people feel happy about.